From a788ec22c8f8eacd867ad7c18b1b11adb2da23e8 Mon Sep 17 00:00:00 2001 From: Ryan Haasken Date: Fri, 13 Dec 2013 18:46:36 -0600 Subject: [PATCH] LUDOC-210 jobstats: Clarify jobstats section and add ALPS Clarified the jobstats section of the documentation by separating the environment variable values for jobid_var from the special values for jobid_var. The special values are now explained in paragraphs with examples following the table. The table's columns were swapped because it is really the job scheduler which determines the value of jobid_var. The description preceding the table was clarified. A run-on sentence was split up at the beginning of section 12.2.1, and the article "a" was added before client. Signed-off-by: Ryan Haasken Change-Id: I9f92891ad44ab3e86d81042d0978d32d9603ecfd Reviewed-on: http://review.whamcloud.com/8577 Tested-by: Jenkins Reviewed-by: Patrick Farrell Reviewed-by: Chris Horn Reviewed-by: Richard Henwood --- LustreMonitoring.xml | 47 +++++++++++++++++++++++------------------------ 1 file changed, 23 insertions(+), 24 deletions(-) diff --git a/LustreMonitoring.xml b/LustreMonitoring.xml index 9fb5055..6137e71 100644 --- a/LustreMonitoring.xml +++ b/LustreMonitoring.xml @@ -340,7 +340,7 @@ Lustre Jobstats
<indexterm><primary>monitoring</primary><secondary>jobstats</secondary></indexterm> Enable/Disable Jobstats - Jobstats are disabled by default, the current state of jobstats can be verified by checking lctl get_param jobid_var on client: + Jobstats are disabled by default. The current state of jobstats can be verified by checking lctl get_param jobid_var on a client: $ lctl get_param jobid_var jobid_var=disable @@ -352,7 +352,9 @@ jobid_var=disable job ID on each client. To permanently enable jobstats on the testfs file system: $ lctl conf_param testfs.sys.jobid_var=SLURM_JOB_ID - The value of jobid_var can be: + The following table shows the environment variables which are set by various job schedulers. + Set jobid_var to the value for your job scheduler to collect statistics on a + per job basis. @@ -360,75 +362,72 @@ jobid_var=disable - Value + Job Scheduler - Job Scheduler + Environment Variable - SLURM_JOB_ID - - Simple Linux Utility for Resource Management (SLURM) - - - - JOB_ID - - Sun Grid Engine (SGE) + SLURM_JOB_ID - LSB_JOBID + Sun Grid Engine (SGE) - Load Sharing Facility (LSF) + JOB_ID - LOADL_STEP_ID + Load Sharing Facility (LSF) - Loadleveler + LSB_JOBID - PBS_JOBID + Loadleveler - Portable Batch Scheduler (PBS)/MAUI + LOADL_STEP_ID - procname_uid + Portable Batch Scheduler (PBS)/MAUI - process name and user ID (for debugging, or if no job scheduler is in use) + PBS_JOBID - disable + Cray Application Level Placement Scheduler (ALPS) - disable jobstats + ALPS_APP_ID - To disable jobstats specify the jobid_var as disable: + There are two special values for jobid_var: disable + and procname_uid. To disable jobstats, specify jobid_var + as disable: $ lctl conf_param testfs.sys.jobid_var=disable + To track job stats per process name and user ID (for debugging, or if no job scheduler is in use), + specify jobid_var as procname_uid: + $ lctl conf_param testfs.sys.jobid_var=procname_uid
<indexterm><primary>monitoring</primary><secondary>jobstats</secondary></indexterm> -- 1.8.3.1